Exosomes Pigmentation

ARTICLE

Received 18 Dec 2014 | Accepted 15 May 2015 | Published 24 Jun 2015

Exosomes released by keratinocytes modulatemelanocyte pigmentationAlessandra Lo Cicero1,2,3, Cedric Delevoye1,2, Floriane Gilles-Marsens1,2, Damarys Loew4, Florent Dingli4,

Christelle Guere5, Nathalie Andre5, Katell Vie5, Guillaume van Niel1,2,3 & Graca Raposo1,2,3

Cells secrete extracellular vesicles (EVs), exosomes and microvesicles, which transfer

proteins, lipids and RNAs to regulate recipient cell functions. Skin pigmentation relies on a

tight dialogue between keratinocytes and melanocytes in the epidermis. Here we report that

exosomes secreted by keratinocytes enhance melanin synthesis by increasing both the

expression and activity of melanosomal proteins. Furthermore, we show that the function of

keratinocyte-derived exosomes is phototype-dependent and is modulated by ultraviolet B.

In sum, this study uncovers an important physiological function for exosomes in human

pigmentation and opens new avenues in our understanding of how pigmentation is regulated

by intercellular communication in both healthy and diseased states.

DOI: 10.1038/ncomms8506 OPEN

1 Institut Curie, PSL Research University, UMR144, CNRS, F-75248 Paris, France. 2 Structure and Membrane Compartments, Centre National de la RechercheScientifique, UMR144, Paris F-75248, France. 3 Cell and Tissue Imaging Facility, Infrastructures en Biologie Sante et Agronomie (IBiSA), Paris F-75248,France. 4 Institut Curie, Centre de Recherche, Laboratoire de Spectrometrie de Masse Proteomique, Paris F-75248, France. 5 Laboratoires Clarins—31chaussee Jules Cesar, Pontoise 95300, France. Correspondence and requests for materials should be addressed to G.R. (email: [email protected]).

NATURE COMMUNICATIONS | 6:7506 | DOI: 10.1038/ncomms8506 | www.nature.com/naturecommunications 1

& 2015 Macmillan Publishers Limited. All rights reserved.

mailto:[email protected]

http://www.nature.com/naturecommunications

In the thin outermost layer of the skin, melanocytes andsurrounding keratinocytes form the epidermal–melanin unit1.Solar irradiation activates signalling cascades that induce the

secretion of molecules including hormones and growth factorsthat lead to increased melanin synthesis in melanocytes2–5. Skinpigmentation requires close intercellular communication andresults in skin tanning but also constitutes an important defensemechanism for photoprotection against Ultraviolet B exposure.Cells communicate via either soluble, secreted factors or viamembrane vesicles, commonly called extracellular vesicles(EVs)6,7. Exosomes are endosome-derived EVs and correspondto the intraluminal vesicles (ILVs) released into the extracellularenvironment on fusion of multivesicular bodies (MVBs) with theplasma membrane. Exosomes harbour membrane and cytosoliccomponents such as proteins, lipids and RNAs8,9. In this study weshow for the first time that, in addition to soluble factors1,10–12,normal human keratinocytes (NHK) release exosomes that play arole in the regulation of pigmentation. Exosomes carryingselected microRNAs (miRNAs) are targeted to melanocytes andmodulate the pigmented status of melanocytes by altering geneexpression and enzyme activity.

ResultsMVBs polarize to intercellular contact sites. Previous studiesreported that keratinocytes secrete vesicles with exosome-likefeatures corresponding to the ILVs of MVBs13. Therefore,MVBs destined for secretion would be found in close proximityto the keratinocyte plasma membrane, as observed in othercell systems14. To visualize MVBs, NHKs were transducedwith a lentivirus vector encoding CD63-GFP, a tetraspaninhighly enriched in MVBs of most cell types15. After 3 daysof transduction, immunofluorescence microscopy (IFM) showedthat CD63-GFP-labelled compartments were primarily distributedaround the nucleus (Fig. 1a, left panel). Interestingly, whentransduced NHKs were co-cultured with normal humanmelanocytes a large fraction of CD63-positive compartmentsredistributed in NHK towards the areas of contact withmelanocytes (Fig. 1a, right panel) as quantified by the increased

distance of CD63-positive compartments from nuclei and relativeto the control (Fig. 1b; Po0.01, t-test). Polarization of MVBs tothe areas of cell–cell contact has similarly been observed in otherexosome-secreting cell systems such as B and T cells14. Such apolarization is specific to MVBs since the distribution of earlyendosomes (EEA1) or Golgi (TGN46) was not drasticallymodified (Supplementary Fig. 1). To get further insight intosuch MVB redistribution in skin models, we analysed cell–cellcontacts in the reconstructed skin epidermis (human epidermalmodel consisting of melanocytes and keratinocytes; see Methods)using immunoelectron microscopy (IEM). At high resolution,these observations revealed CD63-positive MVBs (labelling is forendogenous CD63) close to keratinocyte plasma membranes inthe areas of contact with melanocytes (Fig. 1c), suggesting thatthose could potentially correspond to secretory MVBs.

Keratinocytes secrete exosomes interacting with melanocytes.We undertook a detailed characterization of EVs isolated fromNHK medium with differential centrifugation. Supernatantscontained 30- to 50-nm diameter vesicles partially labelled forCD63 as analysed using IEM (Fig. 2a and Supplementary Fig. 2a).The size of the vesicles observed using IEM is compatible withthat reported for exosomes9 and western blot (WB) analysisrevealed an enrichment for exosomal components in the EVfraction16 such as Alix, CD63 and Tsg101 when compared withwhole-cell lysate (Fig. 2b). To control for the presence of potentialcontaminants, from cell organelles of non-endosomal originor protein aggregates that can co-sediment during ultra-centrifugation, the EV fraction was analysed by WB afterisolation on an iodoxanol (OptiPrep) density gradient17. Alix-and Tsg101-positive vesicles secreted by keratinocytes were onlypresent in the 1.1-g ml� l fraction (Fig. 2c). CD63 was detected inthree consecutive fractions (1.072–1.107 g ml� 1; Fig. 2c) similarto what has been reported for exosomes released by other celltypes17, indicating that the EV fraction consists mostly of ahomogenous population of vesicles with features of exosomes.Reinforcing the exosomal nature of the vesicles secreted by NHK,the proteomic profile of the pellet after the last step of

KeratinocytesCoculture

Dis

tanc

e fr

omnu

clei

(µm

)

6

5

4

3

2

1

0

Keratinocyte

Melanocyte

***TUBULINCD63-GFP

TUBULINCD63-GFPPMEL

CD63-PAG10

MVB

MVB

1 µm

*

*

*

*

*

*

*a b

c

KM

500 nmKeratinocyte

CD63-PAG10 Melanocyte

Figure 1 | MVB polarization in cell culture and reconstructed epidermis. (a) CD63-GFP-transduced NHKs (green) in mono- or co-culture (ratio 1:1,

melanocytes incubated with the same number of keratinocytes) with melanocytes were stained for tubulin (red) and PMEL (melanocyte-specific protein;

blue) and were analysed using IFM (scale bar, 10mm). Asterisks show the nuclei of keratinocytes and the arrow shows the site of contact between

melanocyte and keratinocyte. (b) The distance of CD63-positive compartments from the centre of the corresponding nucleus was quantified in

CD63-GFP-transduced NHK in mono- or co-culture with melanocytes (n¼ 11; ***Po0.01, t-test). (c) EM analysis on ultrathin cryosections of Caucasian-

reconstructed epidermis immunogold-labelled for endogenous CD63 (PAG 10 nm; scale bar, 1 mm). On the right, an inset corresponding to the magnified

area of the back-boxed region depicts an MVB apposed to the keratinocyte plasma membrane in close association with melanocyte.

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8506

2 NATURE COMMUNICATIONS | 6:7506 | DOI: 10.1038/ncomms8506 | www.nature.com/naturecommunications



centrifugation revealed the presence of proteins that have beenreported to be commonly present in exosomes from very differentcell types, such as, for example, major histocompatibility complexclass I, CD9, CD81 and Hsc70, but also cell-type-specific proteinssuch as keratins (Supplementary Data 1). Exosomes can transfernot only proteins but also miRNAs to target cells18,19. Weanalysed the miRNA profile of purified exosomes (SupplementaryData 2). Whereas most identified miRNAs have been previouslyreported to be expressed in NHK20, we also uncovered additionalcandidates (see below) that could be selectively sequestered withinkeratinocyte exosomes. Therefore, our collective results show thatNHKs secrete EVs with features of endosome-derived exosomes,as shown by EM, WB, mass spectrometry and miR profiling9.

We hypothesized that keratinocyte exosomes are targeted tomelanocytes to possibly modulate pigmentation. We tested theability of NHK exosomes to interact with melanocytes byincubating these cells with PKH67-labelled exosomes (Fig. 2e).Exosomes not only interact with melanocytes but are internalizedafter 24 h of incubation as shown by Z projections of theimaged cells. We have also labelled exosomes with fluoresceinisothiocyanate (FITC), and as analysed by fluorescence-activatedcell sorting (FACS), a large majority of melanocytes (95±3%;mean±s.d. throughout) were FITC-labelled compared with thecontrol (Fig. 2e), reinforcing the results that the NHK-isolatedexosomes interact with melanocytes. In addition, NHK-derivedCD63- green fluorescent protein (GFP)-positive exosomes,corresponding to a homogeneous population of vesicles(Supplementary Fig. 2c), were incubated with melanocytes. They

were observed as punctate structures in contact with melanocytes(Supplementary Fig. 2d). Consequently, our data indicate thatkeratinocytes secrete exosomes that interact and are taken up bymelanocytes.

Exosomes regulate melanin synthesis by melanocytes. Toinvestigate the functional effects of NHK exosomes, we isolatedexosomes from Caucasian (low phototype) and Black (highphototype) NHK, which could have potential different effectson pigmentation as the distinct Fitzpatrick phototypes have adifferent capacity to modulate melanin production21; we theninvestigated their effect on pigmentation. The intracellularmelanin content (measured as optical density (OD) at 492 nm)of Caucasian melanocytes incubated with NHK exosomes fromthe same donor or the corresponding medium depleted ofexosomes (supernatant of the last step of exosome purification;Fig. 3a) was not significantly changed in either of the conditionsas compared with the control. In the skin epidermis, solarultraviolet radiation stimulates the production of melanin22, andwe tested whether NHK exosomes released after the ultraviolet Btreatment of cells had an impact on pigmentation. Melanocytesincubated with exosome-depleted medium from ultravioletB-irradiated NHK, which still contain soluble factors, increasedtheir melanin content (E2-fold). These observations corroboratewith previous reports23 showing that ultraviolet B induces thesecretion of soluble factors that stimulate melanogenesis.However, it is important to note that melanocytes incubated

NHK

ExoL

Alix

CD63

Tsg101

96 kDa

75 kDa

50 kDa

37 kDa

46 kDa50 kDa

46 kDa

96 kDa

CD63

Tsg101

Alix

1.06

61.

0721.

0951.

1071.

1141.

1231.

1301.

1351.

1371.

151

1.07

51.

082

1.10

1.12

1.13

1.14

1.15

1.16

1.17

1.21

PBS-PKH Exosomes NHK-PKH

+ ExosomesFITC+92.7

FITC+4.21

FITC

SS

C

Front view

Top viewSid

e vi

ew

100 nm

TYRP1PKHDAPI

TYRP1PKHDAPI

CD63-PAG10

a b c

d

e

Figure 2 | EV characterization and uptake. (a) EM analysis of exosomes from NHK immunogold-labelled for endogenous CD63 (PAG 10 nm, arrows).

(b) WB analysis of exosomes (Exo; 10mg) and total cell lysate (L; 20mg) from NHK or fractions recovered after OptiPrep gradient (c) using anti-Alix,

-CD63 and -Tsg101 antibodies. (d) Analysis by IFM of the interaction of PKH67-labelled (green) exosomes from NHK with melanocytes labelled for

TYRP1 (red) and DAPI (blue; scale bar, 10mm). (e) FACS analysis of melanocytes incubated for 1 h at 37 �C with FITC-labelled exosomes from NHK.

Melanocytes incubated with the last wash of the FITC-labelling were used as a control.

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms8506 ARTICLE




with exosomes secreted by ultraviolet B-treated NHK showed anincreased melanin content as compared with melanocytesincubated with exosomes from non ultraviolet B-treated cells(43±22% and 2±11%, respectively; Po0.05, t-test). Theseobservations lead us to evaluate whether such increased effectscould be due to the secretion of increased amounts of exosomes.However, this was not the case as ultraviolet B exposure of NHKsdid not affect the number of MVBs (Supplementary Fig. 3a) orthe level of total protein in the exosomal pellet (SupplementaryFig. 3b). One hypothesis is that ultraviolet B treatment altered theexosome composition instead, which may then be required forthe modulation of pigmentation. To further test the functionaleffects of exosomes on melanocytes, we isolated exosomes fromBlack NHKs. Black NHK secreted an equivalent amount ofexosome-associated proteins as compared with Caucasian NHKs(Supplementary Fig. 3c). As observed with ultravioletB-stimulated exosomes, the incubation of exosomes releasedfrom Black NHK (Fig. 3a, right plot, second bar), but not thecorresponding depleted medium (Fig. 3a, right plot, fourth bar),increased the melanin content of Caucasian melanocytes(27±13% as compared with normalized control; Po0.05,t-test). Very interestingly, this indicated that secreted exosomesfrom high phototype NHK might intrinsically and without UVstimulation bear components that modulate melanin production.As a control we evaluated whether exosomes from another celltype could induce effects similar to ultraviolet B-stimulated NHKand Black NHK-derived exosomes, which was not the case(Supplementary Fig. 3d). Of note, in our assays melanocytes wereincubated with exosomes recovered from a similar amount ofkeratinocytes. Extrapolated to the skin this ratio is far below theratio of melanocytes and keratinocytes present in the epidermis

(ratio 1:36). Therefore, the effects of exosomes in vivo could beexpected to be even higher.

In order to further investigate their functional properties, weanalysed the ability of exosomes to modulate the activity oftyrosinase (TYR), the key enzyme in melanin biosynthesis23.When Caucasian melanocytes were incubated with NHKexosomes from the same donor, TYR enzymatic activity did notsignificantly increase as compared with cells incubated withoutexosomes (Fig. 3b, second bar). However, a significant increase inTYR activity (150±30%) was observed when Caucasianmelanocytes were incubated with ultraviolet B-stimulatedexosomes from the same donor (Fig. 3b, third bar; Po0.01,t-test), further supporting a model where ultraviolet B stimulatesthe ability of exosomes to increase melanocyte pigmentation.Similarly, a 23±6% increase in TYR activity was observed whenCaucasian melanocytes were incubated with exosomes from BlackNHK (Fig. 3b, fourth bar; Po0.02, t-test), suggesting the effectsof black keratinocytes in the stimulation of pigmentation viaexosomes.

Given that exosomes increase pigmentation and TYR activity,we set out to examine whether they also affect expression ofselected pigmentation genes such as microphthalmia-associatedtranscription factor (MITF), a master transcriptional melanogen-esis regulator24,25. When Caucasian melanocytes were incubatedwith NHK exosomes from the same donor, no significant increase(10±6%) in the expression of MITF-M (the melanocyte isoform)was detected (Fig. 3c) correlating in agreement with theobservation that no increase in pigmentation was observed(Fig. 3a). Accordingly, the expression of MITF-dependent genes,such as TYR and Rab27a, the latter encoding a GTPase involvedin the mobilization of melanosomes to the cell periphery

1.8 1.61.41.2

10.80.60.40.2

0

1.61.41.2

10.80.60.40.2

0M

elan

in c

onte

nt(n

orm

aliz

ed to

con

trol

)in

rec

onst

ruct

ed e

pide

rmis

Mel

anin

con

tent

(nor

mal

ized

to c

ontr

ol)

in r

econ

stru

cted

epi

derm

is

ControlBlack

Exosomes Exosomes

**

22.2

1.81.61.41.2

10.80.60.40.2

0

Mel

anin

con

tent

(nor

mal

ized

to c

ontr

ol)

Tyro

sina

se a

ctiv

ity(n

orm

aliz

ed to

con

trol

)

ControlCaucasianCaucasian UVBBlack

Exosomes Medium w/oexosomes

Exosomes

3

2.5

2

1.5

1

0.5

0

*****

***

**

a b

0.8

0.6

0.4

0.2

0MITF-M Tyr Rab27a

1

1.2

1.4

Rel

ativ

e ge

ne e

xpre

ssio

n(n

orm

aliz

ed to

con

trol

)

* * ****c d e



ControlCaucasianCaucasian UVB

Figure 3 | Exosomal effects on the regulation of pigmentation. (a) Analysis of melanin content (optical density at 492 nm) in Caucasian melanocytes

incubated for 96 h with NHK exosomes (corresponding to 15 mg of protein) resuspended in PBS (left) and medium depleted of NHK exosomes (right;

Caucasian, Caucasian irradiated with ultraviolet B and Black; ratio 1:1, melanocytes incubated with exosomes or medium isolated from the same number of

keratinocytes). Melanocytes incubated without exosomes were used as a control. (b,c) Tyrosinase activity (b) or relative gene expression of Mitf-M,

Tyrosinase and Rab27a (c) were measured in Caucasian melanocytes incubated for 96 h with exosomes from NHK (Caucasian, Caucasian irradiated with

ultraviolet B and Black; ratio 1:1, 15mg of exosomes). Melanocytes incubated without exosomes were used as a control. Intracellular melanin content

analysis of light phototype-reconstructed epidermis incubated with exosomes from Caucasian, ultraviolet B-irradiated Caucasian (d) or Black (e) NHK.

Reconstructed epidermis cultured only with medium and PBS were used as controls. Experiments in d,e were performed with different epidermis of the

same phototype using different melanocyte donors. The data are from three independent experiments. Values are mean±s.d. (*Po0.05; **Po0.02;

***Po0.01, t-test, n¼ 3).





necessary for their transfer to keratinocytes26, was alsounchanged (Fig. 3c). However, a significant increase in theexpression of MITF-M (24±9%; Po0.05, t-test), TYR (26±9%;Po0.05, t-test) and Rab27a (22±10%) was observed inmelanocytes incubated with ultraviolet B-treated exosomes fromthe same donor. The same observations were made withexosomes from Black donors (MITF, þ 22±4% (Po0.05);TYR, þ 26±6% (Po0.02); Rab27a, þ 28±2% (Po0.01);t-test; Fig. 3c). Together, these results indicate on one hand thatultraviolet B irradiation likely induces modifications in thecomposition of exosomes from Caucasian NHK, which thenacquire the ability to stimulate pigmentation. On the other hand,exosomes from Black NHK already contain components thatstimulate the pigmented phenotype of Caucasian melanocytes.

To further probe the pro-melanogenic effects of exosomes in amore physiological context, human-reconstructed pigmentedepidermis was incubated with exosomes isolated from the cellculture supernatants of NHK. Reinforcing our previous resultswith cultured melanocytes, a significant increase in melanincontent of the epidermis was observed with exosomes isolatedfrom cell culture supernatants of ultraviolet B-irradiatedCaucasian NHK (57±13%; Fig. 3d; Po0.05, t-test) and BlackNHK (38±4%; Fig. 3e; Po0.05, t-test) when compared with thecontrol epidermis (4±12%).

Exosomes carry miRNA that regulate pigmentation. Theobservations that miRNAs are possible modulators of pigmen-tation in melanocytes27–29 and that EVs sequester miRNAs18 ledus to investigate whether the observed effects on melanin contentand gene expression could be mediated by miRNAs containedwithin an exosomal vehicle as suggested above30. As a result, weanalysed the miRNA composition of Black, Caucasian andirradiated Caucasian NHK exosomes (Supplementary Data 3).Of note, the miRNA profile of keratinocyte exosomes is modifiedafter ultraviolet B exposure31. Interestingly, only one miRNA,hsa-miR-3196 for which a precise target has not been reported inthe literature, was differentially expressed between irradiated andnon-irradiated Caucasian NHK exosomes (Fig. 4a). In addition,

more than 30 different miRNAs were differentially expressedbetween Caucasian and Black exosomes (Supplementary Table 1).Of note, miR-203 was highly expressed in Black exosomes(Fig. 4b), and this miR was recently reported to be a key regulatorof melanogenesis in melanoma cells, increasing pigmentation andTYR protein levels32.

To directly evaluate the involvement of these miRNAs inpigmentation, Caucasian melanocytes were transfected witheither pre-miR-3196 or pre-miR-203 (Supplementary Fig. 4a,b).Pre-miR-3196 transfection increased the intracellular melanincontent of Caucasian normal human melanocyte (21.5±6.3% ascompared with the control; Supplementary Fig. 5c; Po0.05, t-test) and the gene expression of MITF-M (395±98.7%; Po0.01,t-test) and of Rab27a (225±6.3%; Po0.01, t-test) without greatlymodifying TYR expression (Supplementary Fig. 5a). Pre-miR-203transfection increased the intracellular melanin content(57.5±17%; Fig. 5d; Po0.01, t-test) and, in contrast to mir-3196, enhanced the expression of TYR (41±6.8%; Po0.01, t-test)and Rab27a (17±4.3%; Po0.01, t-test; Supplementary Fig. 5b).MITF-M gene expression was not modified, suggesting that miR-203 may act on TYR regulation independently of the MITFpathway. Similarly, these results reinforce a role of miR-203 in theupregulation of TYR in primary melanocytes in addition tomelanoma cells as reported32 and reveal for the first time a rolefor miR-3196 in melanogenesis. To further test the impact ofthese exosome-associated miRNAs on the process ofpigmentation, Black and Caucasian ultraviolet B-irradiatedkeratinocytes were treated with anti-miR-203 or anti-miR-3196,respectively, to specifically downregulate their expressions andsubsequent loading into exosomes (Supplementary Fig. 4c,d).Exosomes from Caucasian ultraviolet B-irradiated keratinocytestransfected with anti-miR-3196 lose the ability to upregulatemelanin production in melanocytes (Fig. 4e). Exosomes fromBlack keratinocytes transfected with anti-miR-203 did not induceany significant variation as compared with exosomes from NHKtransfected with anti-miR-NC (Fig. 4f), suggesting that additionalexosome-associated components likely compensate for the lack ofmiR-203 or can even be expressed differently in these conditions.Additional molecules, such as proteins or lipids, might cooperate

Caucasian

3

2.5

2

1.5

1

0.5

0

Caucasian UVB

Fol

d ch

ange

miR-3196

Fol

d ch

ange

21.81.61.41.2

10.80.60.40.2

0miR-203a

BlackCaucasian

Mel

anin

con

ten

(nor

mal

ized

to c

ontr

ol)

1.4

1.2

1

0.8

0.6

0.4

0.2

0

Pre-miR-NCPre-miR-3196

**

Control

Caucasian UVB Anti-miR-NCCaucasian UVB Anti-miR-3196

Mel

anin

con

tent

(nor

mal

ized

to c

ontr

ol) 1.6

1.41.2

10.80.60.40.2

0Exosomes

*** **

*

Mel

anin

con

tent

(nor

mal

ized

to c

ontr

ol)

1.81.61.41.2

10.80.60.40.2

0Exosomes

ControlBlack Anti-miR-NCBlack Anti-miR-203

*

21.81.61.41.2

10.80.60.4

NC Pre-miR-NC

Pre-miR-203

0.20

Mel

anin

con

tent

(nor

mal

ized

to c

ontr

ol) ***

a

d e f

b c

Figure 4 | miRNA and pigmentation. (a) Fold change of miR-3196 in exosomes from Black NHK and normalized to exosomes from Caucasian NHK.

(b) Fold change of miR-203-a in exosomes from Caucasian-irradiated NHK and normalized to exosomes from Caucasian NHK. (c–f) Analysis of

intracellular melanin content in Caucasian melanocytes transfected with pre-miR-3196, (c) pre-miR-203 (d) or incubated for 96 h with exosomes from

ultraviolet B-irradiated Caucasian NHK transfected with anti-miR-3196 (e) or with exosomes from Black NHK transfected with anti-miR-203. (f) Values are

mean±s.d. (*Po0.05; **Po0.02; ***Po0.01, t-test, n¼ 3).





with the identified miRNAs and be partially responsible for theobserved effects. In addition, we cannot rule out that suchmolecules might control other properties of exosomes, suchas exosomes from Caucasian ultraviolet B-irradiated NHKpre-treated with trypsin, which failed to increase pigmentation(Supplementary Fig. 6). In sum, our results highlight thatkeratinocyte exosomes modulate their miRNA content to fine-tune melanocyte pigmentation via different signalling pathways(that is, miR-3196 and MITF-dependent or miR-203 andMITF-independent).

DiscussionIn this study we show for the first time that keratinocytescommunicate with melanocytes via EVs with features ofexosomes that carry miRNAs with the capacity to modulatepigmentation.

Keratinocytes have been previously reported to secreteexosomes containing stratifin that stimulate the activity of themetalloprotease MMP-1 in fibroblasts13. Our results furtherexpand the characterization of the EVs secreted by keratinocytes,showing that the vesicles isolated from the cell culturesupernatants of keratinocytes float on a density of 1.1 g ml� l

and express proteins present in the ILVs of MVBs (CD63, Tsg101and Alix). The mass spectrometry analysis identifies proteinsreported to be present in exosomes from different cell types33,also revealing the presence of stratifin as previously shown inkeratinocyte exosomes13. The redistribution of CD63-positiveMVBs towards the sites of contact with melanocytes isreminiscent of the observed polarization of MVBs thataccompanies exosome secretion by immune cells14 as well asother cell types such as neurons and epithelium cells34. CD63-positive MVBs are observed in the human-reconstructedpigmented epidermis at the sites of contact with melanocytes,supporting the exosomal features of the EVs secreted bykeratinocytes. Moreover, we have not observed vesicles largerthan 110 nm in these pellets and we could not detect vesicles orother membranes in the pellet collected during differentialultracentrifugation (10,000g) indicating that keratinocytessecrete primarily small (30–100 nm) membrane vesicles.Although very unlikely, it cannot be ruled out that smallplasma membrane-derived vesicles are also present in thevesicular fraction used in this study.

The EVs/exosomes released by keratinocytes once in contactwith melanocytes induce increased TYR activity as well asexpression of pigmentation genes that logically lead to theobserved increase in melanin content of receiving melanocytes.Our data also highlight that exosomes secreted by keratinocytesfrom different phototypes and after ultraviolet B stimulationcontrol melanocyte pigmentation to a different extent and likelyby distinct mechanisms. Whereas exosomes secreted by Blackkeratinocytes have the ability to stimulate pigmentation to aneven higher extent than the supernatant containing only solublefactors, exosomes secreted by Caucasian keratinocytes do notshow any effect unless the keratinocytes are exposed to ultravioletB at doses close to the physiological solar exposure. Although thesupernatants depleted of exosomes induce melanin production,the exosomes by themselves also increase the pigmentation statusof cells and reconstructed epidermis. The effects of exosomes arecertainly due to their ability to interact with melanocytes and tobe internalized as observed in other cell systems. Such aninteraction will then allow delivery in the endocytic pathway ofcomponents that control the pigmentation process. We show thatexosomes released by the different keratinocytes, Caucasian orBlack, carry a distinct array of miRNAs. Our sequencing analysisreveals that miR-203 is overexpressed in exosomes secreted by

Black keratinocytes. As previously reported, miR-203 targetsKif5b and regulates TYR expression and melanosome transport32,which may explain the observed effects of exosomes. Differentfrom exosomes released by Black keratinocytes and on ultravioletexposure, exosomes from Caucasian keratinocytes show anupregulation of miR-3196. Although this miR shows at least 50targets as predicted by in silico analysis, our study now describesfor the first time its involvement in the pigmentation process.

Collectively, our findings highlight a novel mode of commu-nication between keratinocytes and melanocytes and attribute anovel function for exosomes in the regulation of skin pigmenta-tion. This study also sheds new light on the understanding ofhow the pigmented phenotype is maintained and regulated. Inaddition to soluble factors released by keratinocytes1, exosomes,as membrane-enclosed vesicles carrying membrane proteins andcytosolic components, are likely to participate in the homeostasisof the skin. They could be involved in the modulation/maintenance of the pigmented status of the skin, a process thatcould be altered in disease. These studies open the path for newstrategies to manipulate pigmentation in healthy and diseasedstates as several hyper- and hypopigmentation disorders affectindividuals of all skin phototypes.

MethodsAntibodies and other reagents. Monoclonal antibodies and their sources were asfollows: Rabbit polyclonal anti-b-tubulin (ab59680; 1:500), rabbit anti-EEA1(ab2900; 1:200), horseradish peroxidase (HRP)-conjugated goat polyclonal anti-bodies to rabbit IgG (ab6721; 1:10,000) and to mouse IgG (ab6789; 1:10,000) werefrom Abcam. Mouse monoclonal anti-CD63 (CLB180; 1:200) was from Zymed(Invitrogen). Sheep anti-TGN46 (AHP500GT, 1:400) was from AbD Serotec.Mouse monoclonal anti-TSG101 (GTX70255; 1:500) was from GeneTex. Sheepanti-EGFR (20-ES04; 1:500) was from Fitzgerald. Rabbit anti-TYRP1 (sc-25543;1:200) was from Santa Cruz Biotechnology. Anti-Alix (1:1,000) antibody was a kindgift from Pr. Remy Sadoul. Secondary goat anti-rabbit or anti-mouse antibodiesconjugated to Alexa Fluor-488, -555 or -647 were from Invitrogen (1:200).Protein A-conjugated to 10-nm gold particles was from Cell Microscopy Center(AZU, Utrecht University, Utrecht, the Netherlands).

Cell culture and ultraviolet B treatments. Human foreskin neonatal melanocytesand keratinocytes were obtained from Cellsystems. Keratinocytes were grown inDermaLife Basal Medium with DermaLife K Life Factors. Only keratinocytes fromthe second to fourth passage were used. Melanocytes were grown in DermaLifeBasal Medium with DermaLife M Life Factors. Only melanocytes from the secondto third passage were used. Keratinocytes were seeded and irradiated 24 h later withone shot of 30 mJ cm� 2 of ultraviolet B (312 nm) using a Biosun machine (VilberLourmat, Suarlee, Belgium). Before irradiation, the medium was replaced with PBS,then immediately removed and replaced by the culture medium. Cell viability wasdetermined by the trypan blue (Invitrogen) exclusion test. Keratinocytes wereseeded in six-well plates, incubated overnight and ultraviolet B- or non-irradiatedcells were counted after 24, 48 and 72 h.

Lentiviral transduction of CD63-GFP. Keratinocytes were transduced with thelentiviral vector Cyto-Tracer pCT-CD63-GFP (System Biosciences). After 24 h thecells were washed with PBS and fresh medium was added to them. Conditionedmedium was recovered after 48 h to isolate exosomes.

miRNA transfection. Melanocytes were transfected with pre-miR-203,pre-miR-3196 or pre-miR-NC (as control) using Oligofectamine (Invitrogen). Cellswere transfected a second time after 48 h and analysed after 96 h from the first shotof transfection. Keratinocytes were transfected with anti-miR-203, anti-miR-3196or anti-miR-NC (as control) and conditioned media were recovered after 48 h toisolate exosomes.

Reconstructed epidermis. Reconstructed pigmented epidermis (phototype IV,Sterlab, France) was incubated with growth medium (Sterlab) supplemented withtotal purified keratinocyte-derived exosomes for 2 days or PBS as a control. Then,the medium was replaced daily for 3 days. The epidermis were removed from theirinserts and incubated for 1 h at 100 �C in 400 ml of SOLVABLE (Perkin Elmer) todigest the tissue. Melanin was quantitated on half of the sample by reading its ODat 490 nm. Results were normalized to the control, averaged and they represent amean and s.d. of three independent samples.





Exosome isolation. Exosomes were prepared from conditioned media for 48 h onsubconfluent cells. Media were centrifuged at 2,000g for 15 min (4 �C) and 4,000g(15 min, 4 �C) to remove debris. Then, the supernatant was centrifuged at 10,000gfor 30 min (4 �C) and the exosomes were collected from the supernatant bycentrifugation at 100,000g for 60 min (4 �C), rotor Ti45. The supernatant after thelast step of 100,000g (depleted media) was used to test the effects of soluble factors.The pellet was resuspended and washed in PBS, pH 7.5 as described35. The amountof proteins was quantified and 1.5� 106 keratinocytes release B15mg of exosomes.

For the OptiPrep density gradient, a discontinuous iodixanol gradient wasprepared, 40, 20 and 10% from a stock solution (Sigma) with 0.25 M sucrose,10 mM Tris pH 8 and 1 mM EDTA. The sample was overlaid on the bottom of thegradient and centrifugation was performed at 350,000g for 60 min at 4 �C, rotorSW41. Ten fractions of 490 ml each were collected. The fractions were diluted withPBS, centrifuged at 100,000g for 60 min, resuspended in 30 ml of PBS and stockedat � 20 �C.

To value the role of proteins, exosomes were treated with 0.25% trypsin (Sigma)for 20 min at 37 �C and trypsin was then inhibited (Soybean trypsin inhibitor,Sigma) and washed with PBS.

WB analysis. Cells were lysed on ice in lysis buffer (20 mM Tris, 150 mM NaCl,0.1% Triton X-100, 1 mM EDTA, pH 7.2) with a protease inhibitor cocktail(Roche). Lysates or exosomes were incubated in a sample buffer with or without350 mM 2-mercaptoethanol (Sigma), boiled for 5 min and fractionated withSDS–PAGE using Nupage (4–12%) Bis-Tris gels (Invitrogen) and transferred tonitrocellulose membranes (Millipore). The membranes were blocked in PBS/Tween0.1% (PBS/T) with 5% non-fat dried milk, incubated with the indicated primaryantibody diluted in PBS/T, washed four times in blocking solution and incubatedwith HRP-conjugated secondary antibody followed by washing in PBS/T.Blots were developed using the ECL Plus Western blotting detection system(GE Healthcare) according to the manufacturer’s instruction.

Proteomics. In-gel digestion: Proteins were migrated on a SDS–PAGE gel (10%).Briefly, following SDS–PAGE and washing of the excised gel slices, proteins werereduced with 10 mM dithiothreitol before alkylation with 55 mM iodoacetamide.After washing and shrinking the gel pieces with 100% acetonitrile, in-gel digestionwas performed using trypsin (Sequencing Grade, Promega) overnight in 25 mMammonium bicarbonate at 30 �C.

Liquid chromatography–MS/MS analysis: The extracted peptides were analysedwith nano-LC-MS/MS using an Ultimate3000 system (Dionex S.A.) coupled withan LTQ-Orbitrap mass spectrometer (Thermo Fisher Scientific, Bremen,Germany). Samples were loaded on a C18 precolumn (300 mm innerdiameter� 5 mm; Dionex) at 20 ml min� 1 in 5% acetonitrile, 0.1% TFA. After3 min of desalting, the precolumn was switched online with the analytical C18column (75mm inner diameter� 15 cm; C18 PepMapTM, Dionex) equilibrated in95% solvent A and 5% solvent B (5% acetonitrile, 0.1% formic acid and 80%acetonitrile, 0.085% formic acid). Bound peptides were eluted using a 5–52%gradient of solvent B for 57 min at a 200-ml min� 1 flow rate. Data-dependentacquisition was performed on the LTQ-Orbitrap mass spectrometer in the positiveion mode. Survey MS scans were acquired in the Orbitrap on the 475-1,200 m/zrange with the resolution set to a value of 60.000. Each scan was recalibrated in realtime by co-injecting an internal standard from ambient air into the C-trap (‘lockmass option’). The five most intense ions per survey scan were selected for CIDfragmentation and the resulting fragments were analysed in the linear trap (LTQ).Target ions already selected for MS/MS were dynamically excluded for 180 s.

Processing of MS data: Data were acquired using the Xcalibur software (version2.0.7) and the resulting spectra were then analysed via the MascotTM Softwarecreated with Proteome Discoverer (version 1.4, Thermo Scientific) using theSwissProt Homo sapiens (human) Protein Database. Carbamidomethylation ofcysteines, oxidation of methionine and protein N-terminal acetylation were set asvariable modifications for all Mascot searches. Specificity of trypsin digestion wasset and two missed cleavage sites were allowed. The mass tolerances in MS andMS/MS were set to 2 p.p.m. and 0.8 Da, respectively. The estimated false discoveryrate of all peptide and protein identifications was less than 1%, by automaticallyfiltering on peptide length, mass error and Mascot score of all peptideidentifications.

Interaction of exosomes with melanocytes. FITC-labelled exosomes wereused to characterize the binding and/or the internalization of exosomes fromkeratinocytes to melanocytes. The Fluoreporter FITC protein labelling kit (F-6434,Molecular Probes) was used to label exosomes, according to the manufacturer’sinstructions. Exosomes were labelled and the excess of FITC-reactive solution wasremoved by rinsing in 1 ml PBS. The final washing supernatant was used as thenegative control in the flow cytometry analysis. Data were acquired on a BD AccuriC6 flow cytometer and analysed in FlowJo.

Purified exosomes resuspended in PBS were labelled with PKH67 (Sigma),as control PBS without exosomes was used. The uptake was performed at 37 �Cfor 24 h by incubating melanocytes with the same number of NHK from whichexosomes were recovered. Cells were then fixed with 2% PFA (paraformaldehyde)and staining with anti-TYRP1.

TYR activity assay. TYR activity was determined as previously described36 withslight modifications. Melanocytes were treated with exosomes from keratinocytes(ratio 1:1, melanocytes incubated with exosomes from the same number ofkeratinocytes) for 96 h and then suspended in phosphate buffer containing 1%Triton X-100. After vortexing, cells were clarified by centrifugation at 13,000 r.p.m.for 10 min at 4 �C. TYR (10ml) substrate L-Dopa (15 mM) was incubated with 90 mlof extract in a 96-well plate for 30 min at 37 �C. The absorbance was read at470 nm.

Melanin assay. Melanocytes were treated with exosomes from keratinocytes ormedium depleted of exosomes (ratio 1:1, melanocytes incubated with exosomes ormedium isolated from the same number of keratinocytes) for 96 h and thendisrupted by sonication in 50 mM Tris-HCl, pH 7.4, 2 mM EDTA, 150 mM NaCl,1 mM dithiothreitol and protease inhibitors. Pigment was pelleted at 20,000g for15 min at 4 �C, rinsed once in ethanol/ether (1:1) and dissolved in 2 M NaOH/20%dimethylsulfoxide at 60 �C. Melanin content was measured as OD at 492 nm(ref. 37).

Quantitative real-time PCR. Total RNA was extracted from the control andtreated cells or exosomes using the RNeasy Mini kit (Qiagen) or mirVana miRNAisolation kit (Ambion). The same amount of cDNA was synthesized using NCodeVILOmiRNA cDNA synthesis kit (Invitrogen) and random primers. Real-timePCR was carried out with the LightCycler 480 (Roche), using the Syber greenfluorescent probe, and normalized using the ribosomal gene S26 (for mRNA) orRNU6 (for miRNA). MITF-M: Fw 50-TCTACCGTCTCTCACTGGATTGG-30 ;Rw 50-GCTTTACCTGCTGCCGTTGG-30 . TYR: Fw 50- TGCCAAGCATCCTATCTTCC-30 ; Rw 50-CCATGTAGGATTCCCGGTTA-30 . S26: Fw 50-CCGTGCCTCCAAGATGACAA-30; Rw 50-CGAATGACGAATTTCTTAATGGCCT-30 . Primersfor Rab27a, miR-203, miR-3196 and RNU6 were from QIAGEN.

Immunofluorescence microscopy. Cells were grown on coverslips at 70% con-fluency and then rinsed in PBS and fixed for 15 min in 4% paraformaldehyde/PBSat room temperature. Fixed cells were washed in PBS and quenched for 10 min inPBS/50 mM glycine, saturated in PBS containing 1 mg ml� l BSA (blocking buffer)and permeabilized in PBS/0.05% saponin/1 mg ml� 1 BSA (incubation buffer, IB).Cells were incubated for 1 h with the primary antibody diluted in IB, washed threetimes in IB and incubated with the corresponding secondary antibodies diluted inIB for 45 min. Coverslips were washed three times with IB and then mounted inDABCO medium and examined on an Eclipse 80i Upright Microscope (Nikon)equipped with a CoolSNAP HQ2 CCD Camera, a Piezo Flexure Objective Scannerand � 100 Plan Apo objective (1.4 numerical aperture CFI (chrome-free infinity)).Images are maximum-intensity z projections of three-dimensional image stacksacquired every 0.2 mm using the Metamorph software (MDS Analytical Technol-ogies, Sunnyvale, CA).

Electron microscopy. For ultrathin cryosectioning and immunogold labelling, thecells were fixed with 2% PFA or with a mixture of 2% PFA and 0.2% glutaraldehydein 0.1 M phosphate buffer, pH 7.4. Cells were processed for ultracryomicrotomy(70 nm of thickness), immunolabelled with anti-CD63 and immunogold-labelledusing PAG10 as reported38. For EM of the isolated exosomes, a drop of exosomessuspended in PBS was deposited on Formvar-carbon-coated electron microscopygrids, fixed as above, immunolabelled and stained using the method described forultrathin cryosections38. All samples were analysed using a FEI CM120 electronmicroscope (FEI Company), and digital acquisitions were made with a numericcamera (Keen View; Soft Imaging System, SIS, Germany).

Image analysis and quantification. Quantification of exosome size and number,and MVB number was determined using the iTEM software (Soft Imaging System).

miRNA array profiling. All experiments were conducted at Exiqon Services,Denmark. The quality of the total RNA was verified by an Agilent 2100 Bioanalyzerprofile. Total RNA (140 ng) from sample and reference was labelled with Hy3 andHy5 fluorescent label, respectively, using the miRCURY LNA microRNA Hi-PowerLabeling Kit, Hy3/Hy5 (Exiqon Services) following the procedure described by themanufacturer. The Hy3-labelled samples and a Hy5-labelled reference RNA samplewere mixed pair wise and hybridized to the miRCURY LNA microRNA Array 7th(Exiqon Services), which contains capture probes targeting all miRNAs for human,mouse or rat registered in the miRBASE 18.0. The hybridization was performedaccording to the miRCURY LNA microRNA Array Instruction manual using aTecan HS4800 hybridization station (Tecan, Austria). After hybridization themicroarray slides were scanned and stored in an ozone-free environment (ozonelevel below 2.0 p.p.b.) in order to prevent potential bleaching of the fluorescentdyes. The miRCURY LNA microRNA Array slides were scanned using the AgilentG2565BA Microarray Scanner System (Agilent Technologies Inc., USA) and theimage analysis was carried out using the ImaGene 9.0 software (BioDiscovery Inc.,USA). The quantified signals were background-corrected (Normexp with offsetvalue 10, and normalized using the quantile normalization method, which we have





found produces the best between-slide normalization to minimize the intensity-dependent differences between the samples.

Data availability. The proteomics data set can be downloaded fromhttp://xfer.curie.fr/get/nil/quPNEagSx9J/raw.rar

References1. Yamaguchi, Y., Brenner, M. & Hearing, V. J. The regulation of skin

pigmentation. J. Biol. Chem. 282, 27557–27561 (2007).2. Yamaguchi, Y. et al. Human skin responses to UV radiation: pigment in the

upper epidermis protects against DNA damage in the lower epidermis andfacilitates apoptosis. FASEB J. 20, 1486–1488 (2006).

3. Passeron, T. et al. SOX9 is a key player in ultraviolet B-induced melanocytedifferentiation and pigmentation. Proc. Natl Acad. Sci. USA 104, 13984–13989(2007).

4. Cardinali, G. et al. Melanosome transfer promoted by keratinocyte growthfactor in light and dark skin-derived keratinocytes. J. Invest. Dermatol. 128,558–567 (2008).

5. Raposo, G. & Marks, M. S. Melanosomes--dark organelles enlighten endosomalmembrane transport. Nat. Rev. Mol. Cell Biol. 8, 786–797 (2007).

6. Gyorgy, B. et al. Membrane vesicles, current state-of-the-art: emerging role ofextracellular vesicles. Cell Mol. Life Sci. 68, 2667–2688 (2011).

7. Cocucci, E., Racchetti, G. & Meldolesi, J. Shedding microvesicles: artefacts nomore. Trends Cell Biol. 19, 43–51 (2009).

8. Simons, M. & Raposo, G. Exosomes--vesicular carriers for intercellularcommunication. Curr. Opin. Cell Biol. 21, 575–581 (2009).

9. Raposo, G. & Stoorvogel, W. Extracellular vesicles: exosomes, microvesicles,and friends. J. Cell Biol. 200, 373–383 (2013).

10. Hirobe, T. Role of keratinocyte-derived factors involved in regulating theproliferation and differentiation of mammalian epidermal melanocytes.Pigment Cell Res. 18, 2–12 (2005).

11. Hirobe, T. How are proliferation and differentiation of melanocytes regulated?Pigment Cell Melanoma Res. 24, 462–478 (2011).

12. Cui, R. et al. Central role of p53 in the suntan response and pathologichyperpigmentation. Cell 128, 853–864 (2007).

13. Chavez-Munoz, C., Morse, J., Kilani, R. & Ghahary, A. Primary humankeratinocytes externalize stratifin protein via exosomes. J. Cell Biochem. 104,2165–2173 (2008).

14. Mittelbrunn, M. et al. Unidirectional transfer of microRNA-loaded exosomesfrom T cells to antigen-presenting cells. Nat. Commun. 2, 282 (2011).

15. Escola, J. M. et al. Selective enrichment of tetraspan proteins on the internalvesicles of multivesicular endosomes and on exosomes secreted by humanB-lymphocytes. J. Biol. Chem. 273, 20121–20127 (1998).

16. Thery, C. et al. Proteomic analysis of dendritic cell-derived exosomes: a secretedsubcellular compartment distinct from apoptotic vesicles. J. Immunol. 166,7309–7318 (2001).

17. Tauro, B. J. et al. Comparison of ultracentrifugation, density gradientseparation, and immunoaffinity capture methods for isolating human coloncancer cell line LIM1863-derived exosomes. Methods 56, 293–304 (2012).

18. Valadi, H. et al. Exosome-mediated transfer of mRNAs and microRNAs is anovel mechanism of genetic exchange between cells. Nat. Cell Biol. 9, 654–659(2007).

19. Skog, J. et al. Glioblastoma microvesicles transport RNA and proteins thatpromote tumour growth and provide diagnostic biomarkers. Nat. Cell Biol. 10,1470–1476 (2008).

20. Kraemer, A. et al. UVA and UVB irradiation differentially regulatemicroRNA expression in human primary keratinocytes. PLoS ONE 8, e83392(2011).

21. Fitzpatrick, T. B. The validity and practicality of sun-reactive skin types Ithrough VI. Arch. Dermatol. 124, 869–871 (1988).

22. Choi, W. et al. Regulation of human skin pigmentation in situ by repetitive UVexposure: molecular characterization of responses to UVA and/or UVB.J. Invest. Dermatol. 130, 1685–1696 (2010).

23. Hearing, V. J. & Tsukamoto, K. Enzymatic control of pigmentation inmammals. FASEB J. 5, 2902–2909 (1991).

24. Cheli, Y., Ohanna, M., Ballotti, R. & Bertolotto, C. Fifteen-year quest formicrophthalmia-associated transcription factor target genes. Pigment CellMelanoma Res. 23, 27–40 (2010).

25. Shibahara, S. et al. Regulation of pigment cell-specific gene expression by MITF.Pigment Cell Res. 13(Suppl 8): 98–102 (2000).

26. Van Den Bossche, K., Naeyaert, J. M. & Lambert, J. The quest for themechanism of melanin transfer. Traffic 7, 769–778 (2006).

27. Dynoodt, P. et al. Identification of miR-145 as a key regulator of the pigmentaryprocess. J. Invest. Dermatol. 133, 201–209 (2013).

28. Bell, R. E. & Levy, C. The three M’s: melanoma, microphthalmia-associatedtranscription factor and microRNA. Pigment Cell Melanoma Res. 24,1088–1106 (2011).

29. Dong, C. et al. Coat color determination by miR-137 mediated down-regulationof microphthalmia-associated transcription factor in a mouse model. RNA 18,1679–1686 (2012).

30. Kim, N. H. et al. Reduced MiR-675 in Exosome in H19 RNA-RelatedMelanogenesis via MITF as a Direct Target. J. Invest. Dermatol. 134, 1075–1082(2014).

31. Zhou, B. R. et al. Characterization of the miRNA profile in UVB-irradiatednormal human keratinocytes. Exp. Dermatol. 21, 317–319 (2012).

32. Noguchi, S. et al. MicroRNA-203 regulates melanosome transport andtyrosinase expression in melanoma cells by targeting kinesin superfamilyprotein 5b. J. Invest. Dermatol. 134, 461–469 (2013).

33. Colombo, M., Raposo, G. & Thery, C. Biogenesis, secretion, and intercellularinteractions of exosomes and other extracellular vesicles. Annu. Rev. Cell Dev.Biol. 30, 255–289 (2014).

34. Mittelbrunn, M., Vicente Manzanares, M. & Sanchez-Madrid, F. Organizingpolarized delivery of exosomes at synapses. Traffic 16, 327–337 (2015).

35. Thery, C., Amigorena, S., Raposo, G. & Clayton, A. Isolation andcharacterization of exosomes from cell culture supernatants and biologicalfluids. Curr. Protoc. Cell Biol. Chapter 3, Unit 3–22 (2006).

36. Busca, R., Bertolotto, C., Ortonne, J. P. & Ballotti, R. Inhibition of thephosphatidylinositol 3-kinase/p70(S6)-kinase pathway induces B16 melanomacell differentiation. J. Biol. Chem. 271, 31824–31830 (1996).

37. Wasmeier, C. et al. Rab38 and Rab32 control post-Golgi trafficking ofmelanogenic enzymes. J. Cell Biol. 175, 271–281 (2006).

38. Raposo, G., Tenza, D., Murphy, D. M., Berson, J. F. & Marks, M. S. Distinctprotein sorting and localization to premelanosomes, melanosomes, andlysosomes in pigmented melanocytic cells. J. Cell Biol. 152, 809–824 (2001).

AcknowledgementsWe are grateful to Eric Gooris for insightful discussions and support during the course ofthis work. We are grateful to Allison de Horsey (St George’s School, Newport, RI, USA)for critical reading and editing of the manuscript. We are grateful to Matteo Gentilifor help with FACS analysis and to Joanna Kowal for the exchange of expertise ontechniques. We thank Lucie Sengmanivong and Vincent Fraisier from the PICT-IBiSALhomond UMR144 Imaging facility of Institut Curie and the Nikon Imaging Centreat Institut Curie-CNRS, members of the France-BioImaging national researchinfrastructure. We acknowledge France-BioImaging infrastructure supported by theFrench National Research Agency (ANR-10-INSB-04, )Investments for the future*).This work was supported by Institut Curie, CNRS and Clarins.

Author contributionsA.L.C., G.R., G.v.N. and C.D. designed the study and analysed data; C.D. performed theanalysis on reconstructed epidermis; F.-G.M. performed the electron microscopy on thereconstructed epidermis; D.L. and F.D. conducted mass spectrometry analysis; A.L.C.carried out all the other experiments; C.G, N.A. and K.V. provided technical assistancefor the ultraviolet irradiation, participated in the scientific discussion and financed theproject; A.L.C. and G.R. wrote the manuscript; C.D. and G.V.N. edited the manuscript;G.R. directed the study.

Additional informationSupplementary Information accompanies this paper at http://www.nature.com/naturecommunications

Competing financial interests: The authors declare no competing financial interests.

Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/

How to cite this article: Lo Cicero, A. et al. Exosomes released by keratinocytes mod-ulate melanocyte pigmentation. Nat. Commun. 6:7506 doi: 10.1038/ncomms8506 (2015).

This work is licensed under a Creative Commons Attribution 4.0International License. The images or other third party material in this

article are included in the article’s Creative Commons license, unless indicated otherwisein the credit line; if the material is not included under the Creative Commons license,users will need to obtain permission from the license holder to reproduce the material.To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/




http://xfer.curie.fr/get/nil/quPNEagSx9J/raw.rar



http://npg.nature.com/reprintsandpermissions/

http://npg.nature.com/reprintsandpermissions/

http://creativecommons.org/licenses/by/4.0/


Documents

Exosomes Pigmentation